Revolve seam with spline!

[HughesTooling]

There've been a couple of post complaining about the seam formed by revolve, apparently it shows when 3d printing. What I've noticed is you only get a seam from a spline.

Top curve is a spline, bottom an arc. There's no seam for the arc or line.

The file's attached.

Here's a link to the thread with the problem 3d printing. I would have thought the meshing software should deal with this.

http://forums.autodesk.com/t5/design-validate-document/how-to-get-rid-of-seam-created-by-making-a-bo...

Mark

[PhilProcarioJr]

@HughesTooling

It doesn't leave a seam when 3d printing parts like this using FDM/FFF printers. I have printed hundreds of revolved objects and no seam. The reason they are getting the seam is because of the settings in their 3d print software. This is a known problem in the 3d printing world. This is why most 3d printers have a vase print algorithm. This is also a problem new people encounter because they don't have their printer setup properly. The reason they get a seam is because the slicer decides each layers start and stop point. If the extruder is not extruding enough material you get a gap, if it is extruding too much it leaves extra material sticking outside the model. There are settings to move the start and end points in a layer. There are also settings for minimum and maximum travel between breaks.....etc

This is also one of the largest reasons the big boy rapid prototyping machines use laser sintering, laser solidification and polyjet.

[michallach81 · Accepted solution]

It can't be done any different.

If we revolve arcs or line (polyline) then surface (bodys boundary) is defined by very basic equations (like sphere eqation x²+y²+z²=R², maybe not exactly this one but with same variables, position of center point and length of radius), where if we have more complex geometry we define surface by NURBS surface equation (). NURBS surface is always rectangular (4 sides). Nurbs surface can take form of some revolved shape but it will have a seam where two edges connect.

Real problem is that we can see seam in 3D print but that is caused by approximation of nurbs surface while tessellation is taking place.

[HughesTooling]

Thanks @michallach81 and @PhilProcarioJr I posted more out of curiosity, I noticed the seam on a few problem turned parts in the CAM workspace and just wondered what the difference was.

Mark

[HughesTooling]

Here's something odd, Fusion hiding the seam on revolved arcs can be a problem because there's no way to measure them. I helped someone in the CAM forum who'd imported a stp file and there were no seams so you had to create a sketch and intersect to get sizes. I just tested an export from Rhino that has a seam, when opened in Fusion the seams gone! So is Fusion rebuilding the surface or just hiding the seam?

Rhino.

Imported in Fusion, the hiding of the seam can be a bit inconvenient if you don't know the sizes of the part.

Mark

[PhilProcarioJr]

@HughesTooling

When importing a dumb solid in Fusion or almost any CAD app for that matter it is rebuilding the surfaces from the boundary information. Which is "probably why the seam disappears.

Can you share your file with the seam I want to test something....

[HughesTooling]

STP file attached.

Mark

[michallach81]

Wierd it must be Rhino 5. In theory Rhino build only NURBS for any geometry. From 5th it can create "solids" (prismatic parts), but they are "translated" to NURBS when used (trimmed, exploded). 

I was pretty sure that when translated to STEP, Rhino would write only NURBS, but it looks like it preserved data for prisms to be recreated.

@PhilProcarioJr There is no software that would read NURBS (let say, sphere) and based on that guess what prism it represents. Even tide approximation will always be to big. Resaults would be always a guess. That's why I think Rhino did write additional data.

[PhilProcarioJr]

@HughesTooling

Just as @michallach81 described the difference between a simple surface equation and a complex nurbs equation your model with a curvature map shows the difference in complexity between the top and bottom surfaces. The top is has much more complex curvature then the bottom.

[LMD001]

Hello @HughesTooling,

Had a similar question some time ago, @jeff_strater answered me this at the time: http://forums.autodesk.com/t5/design-validate-document/revolve-ellipse-leaves-profile-behind/m-p/574....

Best regards,

Ludo

[mattad_spam]

Thanks for the assist. I am an old hack with 3D modelling, but very new to 3D printing and trying to learn. Am taking the issue over to the Cura boards to see if they can help.

Appreciate the support.

[Anonymous]

I get why Fusion 360 puts the edge there but is there any way to just hide that single edge?

[PhilProcarioJr]

@Anonymous

There is no way to turn of the display of that single edge.

The only way is to turn your visibility to shaded only but then you lose all edges in the display, but I have to ask what does it mater if the edge is displayed in the model mode?

[Anonymous]

It is not functionally in the way, it just bothers me. Parts look sloppy with extra edges.

[PhilProcarioJr]

@Anonymous

Every CAD app I have ever used does this, Solidworks, Pro-E, Catia...etc it's normal and there is nothing sloppy about it. The surface of the model is fine and that's all that matters.

[Anonymous]

Did you ever find a solution to this?

The seam-line from a revolve is showing up in technical drawings and it is not an edge that should exist/be present in the design.

[jnf4c]

Thanks a lot.  The spline revolve is the culprit.  Been months scratching my head on that one.

Thanks again.

Joe F.